A Comparison of Volume-Staged Gamma Knife Radiosurgery and Hypofractionated Radiotherapy for Treating Large Arteriovenous Malformations

LIJUN MA1, Michael W McDermott1, Nalin Gupta2, Igor Barani3, Shannon Fogh2, Jean Nakamura2, Penny K Sneed, MD4, Arjun Sahgal5, David Larson3

1UCSF 2University of California San Francisco, San Francisco, CA, USA 3San Francisco, CA United States 4Dept. of Radiation Oncology, University of California San Francisco 5University of Toronto

Keywords: arteriovenous malformation, cyberknife, gamma knife, radiotherapy, radiosurgery

Interactive Manuscript

Ask Questions of this Manuscript:



     There are different irradiation strategies for large arteriovenous malformations of the brain.
To compare hypofractionated Cyberknife treatments with volume-staged Gamma Knife treatments for large arteriovenous malformations (AVMs) in terms of sparing delivered dose to normal brain tissue.
     Treatment plans of all 7 pediatric patients treated with volume-staged Gamma Knife Perfexion radiosurgery (mean target volume =14.9 ±3.0 mL) at our institution since 2007 were analyzed. For each patient case, target and normal tissue contours were extracted to recreate a full target volume treatment plan for a hypofractionated Cyberknife treatment on a research license treatment planning system (Multiplan 5.0, Accuray). A reference 12-Gy normal brain volume for each patient was first determined by summing the 12-Gy normal brain volumes from individual volume-staged Gamma Knife treatment plans. Then the peripheral target dose for the final single-fraction Cyberknife plan was adjusted such that it yielded the same 12-Gy normal brain volume as the reference 12-Gy normal brain volume. From this determined peripheral target dose, dose-fractionation schemes were derived via the linear-quadratic model for comparable normal brain (?/?=3) sparing.
     Cyberknife treatments produced identical target volume coverage (95-98%) and normalized conformity indices (nCI=1.24-1.46) as the summed volume-staged Gamma Knife treatment plan for each patient. However, to achieve the same reference 12-Gy normal brain volume for each patient, an average reduction of 18.7% (±7.3%) in the peripheral target dose was required for the Cyberknife treatment plan as compared to the volume-staged Gamma Knife treatments (p<0.001). This led to dose fractionation schemes such as 28 Gy in 5 fractions or 24 Gy in 3 fractions or 14.0 Gy in 1 fraction for the hypofractionated Cyberknife treatments according to the linear-quadratic model.
     When conservatively accounting for the dose to the normal brain, volume-staged Gamma Knife radiosurgery allowed a higher biologically equivalent prescription dose to be delivered compared with full-volume hypofractionated treatments for large AVMs.
     Radiosurgery and hypofractionated irradiation may have different biological effects on AVM tissue.


Project Roles:

L. MA (), M. McDermott (), N. Gupta (), I. Barani (), S. Fogh (), J. Nakamura (), P. Sneed (), A. Sahgal (), D. Larson ()